© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129061
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on literal constants and numeric or boolean values in code (e.g. numbers, `None`, `false`, assignments, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    optional applicable existence

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aucune
    -0.87
     first
    -0.85
     nowa
    -0.85
    ניס
    -0.82
    𝚠
    -0.82
     실패
    -0.81
    cabul
    -0.80
     pierwsze
    -0.79
     firstname
    -0.77
     who
    -0.76
    POSITIVE LOGITS
     additional
    1.07
    ditional
    1.07
     existence
    0.93
     separate
    0.92
    存在的
    0.91
     додат
    0.90
     дополнитель
    0.89
    separate
    0.85
     supplémentaire
    0.85
     optional
    0.85
    Activations Density 0.010%

    No Known Activations